Merge branch 't/oss_error_message_fix'
[paraslash.git] / NEWS
1 ------------------------------------------
2 0.?.? (to be announced) "spectral gravity"
3 ------------------------------------------
4
5 - UTF8 support for para_gui and the mp3 audio format handler.
6 - Scheduler improvements and fixes.
7 - The obsolete gettimeofday() function has been replaced
8 by clock_gettime() on systems which support it.
9 - Speed and usability improvements for para_gui.
10 - para_client now restores the fd flags of stdin and stdout
11 on shutdown
12
13 -----------------------------------------
14 0.4.12 (2012-12-20) "volatile relativity"
15 -----------------------------------------
16 The new command line player, the resample filter, ALSA support for
17 para_fade, and the improved build system are the highlights of this
18 release which probably marks the end of the 0.4.x series.
19
20 - The afh receiver and the para_play executable.
21 - The resample filter: A sample rate converter based on
22 libsamplerate.
23 - The "versions" directory has been removed from the master
24 branch. The tarballs of the old releases are now available
25 in the new "releases" branch.
26 - Overhaul of the build system: All generated files are now
27 written to the "build" directory.
28 - The modular mixer API and the alsa mixer.
29 - Minor fixes for the osx writer.
30
31 --------------------------------------
32 0.4.11 (2012-07-20) "mutual diversity"
33 --------------------------------------
34
35 The major feature in this release is the new sideband API for
36 client-server communication. This API will be used exclusively starting
37 with 0.5.0, which breaks backward compatibility but allows to get rid
38 of quite some compatibility code. Other noteworthy changes include
39 decoder latency improvements and a long-standing bug fix for the
40 ALSA writer.
41
42 - Sideband connections: If both para_server and para_client
43 support this feature, data is sent as a multiplexed stream.
44 - The --no_default_filters option of para_filter has been
45 removed.
46 - Several fixes and latency improvements to various decoders.
47 - The ALSA writer now limits the prebuffer time to 500ms.
48 - Documentation improvements.
49 - Overhaul of the command_util.sh script.
50 - Fixes for some minor problems found by the clang analyzer.
51 - Compiles (almost) without warnings on gcc-3.
52 - Robustness improvements of the buffer tree code.
53
54 ------------------------------------------
55 0.4.10 (2012-03-30) "heterogeneous vacuum"
56 ------------------------------------------
57
58 Nothing earth-shaking in this release, but quite a few usability
59 improvements and the usual mix of cleanups and fixes.
60
61 - The --no_default_filters option of para_filter has been
62 deprecated. It still works but has no effect and will be
63 removed in the next version.
64 - para_gui now prints also the stderr output of the executing
65 command in the bottom window.
66 - Cleanup and consolidation of the various wrappers for
67 write(), writev(), send() and friends.
68 - The obscure error messages on mmap() failures have been
69 replaced by meaningful messages. This affects mainly
70 para_afh.
71 - para_audioc: Cleanups and memory leak fixes.
72 - Test 0004-server no longer fails if para_server is not
73 being built.
74 - New configure options: --with-id3tag-{headers,libs}.
75
76 -------------------------------------
77 0.4.9 (2011-12-06) "hybrid causality"
78 -------------------------------------
79
80 Support for another audio format, interactive mode for para_client
81 and para_audiod and many small improvements/fixes all over the place.
82
83 - Support for flac, the free lossless audio codec.
84 - Fix for an endless loop in the mp3 decoder for certain
85 (corrupt) mp3 files.
86 - When executed without specifying a command, para_client
87 and para_audioc start an interactive shell (requires
88 libreadline being installed). The interactive mode offers
89 full tab completion and command line history.
90 - autogen.sh now detects a distcc setup and adjusts the
91 parameter for the -j option of make accordingly.
92 - Shared memory areas are no longer restricted to 64K. We now
93 detect the maximal size of a shared memory area at runtime.
94 - cleanup of the internal uptime API.
95 - para_server prefaults the mmapped audio file to avoid
96 delays on slow media.
97 - A new test for the test-suite that exercises the
98 communication between para_server and para_audiod.
99 - The alsa writer eats up less CPU cycles when configured to
100 use the DMIX plugin.
101 - Simplified and unified receiver code.
102 - Makefile cleanups.
103 - Commands which print a list of matching audio files now
104 emit a meaningful error message if no audio file matched the
105 given pattern(s).
106
107 --------------------------------------
108 0.4.8 (2011-08-19) "nested assignment"
109 --------------------------------------
110
111 Gcrypt support, the overhauled osx writer and regex format specifiers
112 are the highlights of this release.
113
114 - support for libgcrypt as a drop-in replacement for openssl.
115 Run configure --enable-cryptolib=gcrypt to link against
116 libgcrypt. The two crypto implementations are compatible to
117 each other, i.e. a para_client executable linked against
118 gcrypt can connect to para_server linked against libssl
119 and vice versa.
120 - Rewrite of the osx writer (output plugin for Mac OS).
121 - audiod: The format specifier for receivers, filters and
122 writers is now treated as a regular expression. This allows
123 to replace 5 lines in the config file (one for each audio
124 format) by one single line. See the manual for details.
125 - The *.cmdline.[ch] files are no longer contained in the released
126 tarballs. This reduces the size of the tarballs but requires
127 gengetopt to build the tarball.
128 - Compiles cleanly also with llvm/clang.
129 - Corrupt mp3 files are handled more gracefully.
130 - The alsa writer uses poll fds instead of computing timeouts.
131 - Cleanup of the generic writer API.
132 - sched: Optimized zero timeouts.
133 - vss timeout cleanups.
134 - oggdec fixes and improvements.
135
136 --------------------------------------
137 0.4.7 (2011-06-01) "infinite rollback"
138 --------------------------------------
139
140 The new ao writer, support for ssh RSA keys and a couple of other
141 enhancements.
142
143 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
144 platforms through the libao audio library.
145 - Support for RSA keys generated with ssh-keygen.
146 - configure: improved options for ogg/vorbis/speex.
147 - The git version reported by --version always matches HEAD.
148 - The autogen script detects the number of processors and
149 runs a parallel make if possible.
150 - Major cleanup of the crypto API.
151 - Documentation updates.
152
153 ------------------------------------------
154 0.4.6 (2011-03-31) "deterministic entropy"
155 ------------------------------------------
156
157 Lots of ogg/vorbis improvements, the new test suite, enhancements
158 for para_gui and a fair amount of other bug fixes.
159
160 - For DCCP/OGG streams the audio file header is only sent once
161 at the beginning of the stream rather than periodically
162 every five seconds. This reduces network traffic and the
163 FEC group size.
164 - The vorbis comment header is replaced by an empty dummy header
165 before the header is sent over the network. This also results in
166 less network traffic and smaller FEC groups.
167 - The new "test" make target allows to perform some sanity checks prior
168 to installing the package.
169 - ogg timing fixes and performance improvements
170 - Scheduler improvements
171 - Proper exit codes for para_write
172 - para_gui: New option --theme to select a startup theme. Several
173 other improvements and fixes.
174 - aacdec error message cleanups
175 - simplified color error handling
176
177 --------------------------------------------
178 0.4.5 (2010-12-17) "symmetric randomization"
179 --------------------------------------------
180
181 Bug fixes, internal cleanups and variable-sized FEC slices.
182
183 - Contains a fix for an invalid-free-bug in the ogg audio format
184 handler code.
185 - Switching off the DCCP sender works again.
186 - para_audiod handles crashes of para_server more robustly.
187 - Internal scheduler and writer cleanups.
188 - Reduced latency due to variable-sized FEC slices.
189 - Improved documentation and error diagnostics.
190 - The build of para_server is now optional, allowing the build
191 to succeed in case libosl is not installed.
192
193 ------------------------------------------
194 0.4.4 (2010-08-06) "persistent regularity"
195 ------------------------------------------
196
197 Support for yet another audio format, para_write improvements and
198 bug fixes.
199
200 - Support for the speex codec.
201 - Support for sample formats other than 16 bit little endian.
202 - error2.h is now created by a perl script which speeds up configure
203 considerably.
204 - Fix a bug in the aac decoder which could lead to segfaults in
205 para_filter/para_audiod.
206 - Fixes for autoconf-2.66.
207
208 ----------------------------------------
209 0.4.3 (2010-07-05) "imaginary radiation"
210 ----------------------------------------
211
212 Many improvements for the DCCP and the UDP transport, the new user
213 manual and the usual mix of bug fixes and internal improvements.
214
215 - FEC support for the DCCP sender (Gerrit Renker). The new
216 --dccp_max_slice_size, --dccp_data_slices_per_group and
217 --dccp_slices_per_group options can be used to set the FEC
218 parameters for the DCCP transport.
219 - DNS lookups for UDP targets (Gerrit Renker).
220 - The new user manual replaces the README, README.afs, REQUIREMENTS
221 and INSTALL documents.
222 - Fix an end-of-file detection bug in the oggdec filter.
223 - The new nonblock API.
224 - Both options of the oggdec filter have been removed.
225 - New debug mode for the internal scheduler.
226
227 ------------------------------------------
228 0.4.2 (2010-04-23) "associative expansion"
229 ------------------------------------------
230
231 It's been some time since the last release, but finally here is
232 paraslash-0.4.2. The bulk of the changes comes from the new buffer
233 tree API, but there are changes all over the tree. Mainly performance
234 and usability improvements, but also quite some bug fixes.
235
236 - The new buffer tree API.
237 - DCCP: Support for CCID negotiation (Gerrit Renker).
238 - UDP robustness fixes.
239 - The --bufsize option for mp3dec is gone as it no longer makes sense
240 for the new buffer tree API.
241 - Fix audible buffer underruns for wma streams.
242 - The alsa writer no longer prints meaningless underrun durations.
243 - audiod: Defaults work also for udp streams. If no filter is
244 given for an audio format that is received via upd, fecdec is
245 automatically added as the first filter (along with the decoder).
246
247 ---------------------------------------
248 0.4.1 (2009-12-22) "concurrent horizon"
249 ---------------------------------------
250
251 Support for another audio format, minor feature enhancements and lots of bug
252 fixes. All fixes that have been accumulated in the maint branch (in particular
253 those mentionened in the 0.3.6 release notes) appear in this release as well.
254
255 - wma support.
256 - new afh option: --human to activate human-readable output.
257 - new server/audiod option: --log-timing to print timing information.
258 - build system improvements.
259 - source code documentation updates.
260
261 -------------------------------------
262 0.3.6 (2009-12-07) "cubic continuity"
263 -------------------------------------
264
265 Quite a few bugs have been found and fixed since 0.3.5, so here's
266 another 0.3.x release. No new features.
267
268 - Always check return value of malloc().
269 - ogg vorbis/FEC: Do not write garbage after the audio file header.
270 - exit if root privileges could not be dropped.
271 - FEC: Fix computation of extra slices.
272 - oss: Fix check for empty input buffer.
273 - Avoid buffer underruns due to filter chain output buffer constraints.
274 - server: Fix assignment of afs_pid.
275 - Don't panic if the afs database contains unknown audio formats.
276 - http/dccp: Do not send the audio file header twice.
277 - FEC: Timing improvements.
278
279 ----------------------------------------------
280 0.4.0 (2009-11-10) "simultaneous independence"
281 ----------------------------------------------
282
283 Two significant changes which require the new version number: The
284 improved authentication dialog and the fact that the database code
285 has been moved to a library, libosl. To use the new version, you have
286 to generate new RSA keys, see INSTALL for details. A shell script is
287 provided for conversion of the 0.3 database to the new 0.4 format.
288
289 - stronger crypto for client authentication
290 - the database code has been moved to a library
291 - improved status item handling
292 - cleanup of the build system
293 - The "-V" option now also prints the git version
294 - the new parser-friendly listing mode for the ls and stat commands
295 - mandatory rc4 encryption
296 - major audio format handler cleanups
297 - (id3,...) tags are no longer stored as a combined string in the database
298 - new mood methods: artist_matches, title_matches, comment_matches,
299 album_matches, year_maches, year.
300
301 --------------------------------------------
302 0.3.5 (2009-09-21) "symplectic separability"
303 --------------------------------------------
304
305 Full client support for *BSD Unixes, complete re-write of the ogg
306 vorbis audio format handler, various improvements all over the place
307 and the usual mix of bugfixes. This release marks the end of the 0.3
308 series if no serious problems show up.
309
310 - the new oss writer (supported on *BSD and Linux)
311 - rewrite of the ogg vorbis audio format handler. It's
312 recommended to replace the chunk tables of existing ogg
313 vorbis files in the afs database by re-adding these files
314 with "add -f".
315 - support for netmask subsets (Gerrit Renker)
316 - the new prebuffer filter
317 - improved signal handling
318 - variable fec output buffer size
319 - improved FEC timing fixes audible buffer underruns in UDP mode
320 - --log_color actually works
321 - new ls option: -d (print dates as seconds after the epoch)
322 - update to gengetopt 2.22.2
323 - support for RSA keys of size > 512 bits
324 - new option "mixer_channel" for para_fade
325
326 -----------------------------------------
327 0.3.4 (2009-05-07) "elliptic inheritance"
328 -----------------------------------------
329
330 The new udp sender, forward error correction, colored logs and various
331 other improvements. As the udp sender does not depend on any special
332 libraries, it is built unconditionally.
333
334 - The udp sender replaces the ortp sender. The new code uses forward
335 error correction to protect against packet losses. Many thanks to
336 Gerrit Renker for providing ipv6 support.
337 - The default port for udp streaming now defaults to 8000, like
338 for the http and the dccp senders/receivers.
339 - Loglevels are now specified as symbolic names, e.g.
340 "--loglevel info".
341 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
342 - para_server/para_audiod: Color support for log messages.
343 - new options for mp3dec: --ignore-crc, --bufsize
344 - new audiod option: --config-file.
345 - gengetopt cleanups.
346 - Improved help/man pages: The documentation of para_audiod,
347 para_recv, para_filter and para_write now also contains
348 all options of the available receivers/filters/writers. The
349 man page of para_fade contains a description of the different
350 modes of operation.
351 - More source code documentation.
352 - vss timing fixes.
353
354 --------------------------------------------
355 0.3.3 (2008-12-01) "axiomatic perspectivity"
356 --------------------------------------------
357
358 Internal code cleanups, bug fixes, improved tag handling and the new
359 amplification filter.
360
361 - para_server uses the generic scheduling code.
362 - overhaul of the virtual streaming system.
363 - mp3: id3 version 2 support via libid3tag (optional)
364 - ogg: vorbis comment support.
365 - aac meta info support.
366 - mp3 audio format handler cleanups.
367 - new filter: "amp" to amplify the amplitude of the audio stream
368 - new status item/database entry: amplification. It is
369 used by the amp filter to pre-amplify the audio stream.
370 - fix a close-without-open bug in para_write.
371 - fix a bug in com_init() which was introduced in 0.3.2.
372 - better error diagnostics for para_client.
373
374 -----------------------------------------
375 0.3.2 (2008-04-11) "probabilistic parity"
376 -----------------------------------------
377
378 The new para_afh executable, scheduling and documentation improvements.
379
380 - new ls option: -lc (list chunk table)
381 - new executable: para_afh, the stand-alone audio file handler tool
382 - afs commands can send output more than SHMMAX (32MB on Linux). This
383 also reduces the memory usage of commands that produce large amounts
384 of output.
385 - major scheduler and audiod cleanups.
386 - more detailed and much nicer man pages.
387
388 ---------------------------------------
389 0.3.1 (2008-02-23) "liquid interaction"
390 ---------------------------------------
391
392 A mix of cleanups, bug fixes, improvements, and some new features. No
393 significant changes to the new database (osl) code, which is generally
394 a good sign.
395
396 - Share some similar/duplicate code between the http and the
397 dccp sender.
398 - Generic access control lists for paraslash senders.
399 - dccp sender: Access control lists, connection limiting and support
400 for the allow,deny,on,off,help sender commands.
401 - The default dccp port changed from 5001 to 8000 (suggested by
402 Gerrit Renker).
403 - para_server starts even if not all public keys could be loaded.
404 - Audiod performance improvements.
405 - fix a bug in the "off" command of the http sender.
406 - fix some fd and memory leaks.
407 - Update to gengetopt-2.22.
408
409 -------------------------------------
410 0.3.0 (2008-01-12) "solar saturation"
411 -------------------------------------
412
413 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
414
415
416 Usually one might expect lots of new features AND a big increase in size
417 for a major release like this.
418
419 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
420 ever. The decrease in size is mostly due to the removal of some
421 graphical tools (which were only quick hacks anyway). But also the
422 fact that the mysql code is gone cuts down the size a bit.
423
424 Being independent of mysql comes at a cost: The fact that paraslash
425 now contains its own database (the object storage layer, osl) increases
426 the (stripped) binary size of para_server by ~50K on i386.
427
428 - no more restrictions on unique basenames.
429 - independent of mysql: The new self-contained object
430 storage layer (osl) replaces the mysql database.
431 - New executable para_fsck: Check integrity of osl tables.
432 - Lyrics support.
433 - Reliable audio file move/rename detection.
434 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
435 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
436 Solaris (sparc64).
437 - the new osl-based audio file selector (afs) replaces the random,
438 the playlist and the mysql selector of paraslash-0.2.x.
439 - IPv6 support (thanks to Gerrit Renker).
440 - paraslash-0.2.x streams are now called "moods". Writing
441 0.3.x-mood definitions should be both easier and more
442 powerful than writing 0.2.x-stream definitions.
443 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
444 been removed. The world is a better place without them. However,
445 para_gui is still there.
446 - afs tracks audio file selection also in playlist mode.
447 - few easy-to-use afs commands replace the many not-so-easy-to-use
448 mysql commands (and are available also in playlist mode).
449 - Improved error subsystem.
450 - The earth-shaking new logo.
451
452 -----------------------------------------
453 0.2.17 (2007-11-20) "isotropic threshold"
454 -----------------------------------------
455
456 Mainly bugfixes and cleanups in this version which marks the end of
457 the 0.2.x series if no serious bugs show up after the release.
458
459 - mysql_selector: fix a locking bug.
460 - universal chunk queueing.
461 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
462 to Gerrit Renker).
463 - be more carful wrt. signed vs. unsigned argument passing.
464 - cleanup error.h and fix some references to invalid error
465 codes.
466 - update to gengetopt-2.21.
467 - update to ortp-0.13.1.
468 - autoconf: extend checks for headers, library functions and
469 compiler characteristics.
470 - Fix streaming of large mp3 files.
471 - Fix an off-by-one bug in playlist handling.
472
473 --------------------------------------
474 0.2.16 (2007-04-05) "neural discharge"
475 --------------------------------------
476
477 The main change in this release is the major audio format handler
478 cleanup which removes some similar/duplicate code and makes it easier
479 to implement plugins for other audio formats. Of course, the usual mix
480 of other improvements/changes/bugfixes also made it into the release.
481
482 - simplified audio format handlers (most of the handling functions
483 were moved one layer up to the virtual streaming system).
484 - para_server uses mmap to read audio files
485 - repositioning of mp3 streams is much faster, in particular for
486 jumping near the end of large mp3 files.
487 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
488 and AFS_WRITE.
489 - fix a bug in para_filter that caused decoding of aac files
490 to start only after a few seconds.
491 - fix osx_writer hangs
492 - simplified dccp code (thanks to Gerrit Renker)
493 - the compress filter works also on big endian systems (ppc)
494
495 -----------------------------------------
496 0.2.15 (2007-02-16) "inductive resonance"
497 -----------------------------------------
498
499 Minor improvements, more documentation and a bunch of bug fixes.
500
501 - para_server: The server.users file is only read once on server
502 startup rather than for each connection
503 - mp3dec: Fix decoding of corrupt mp3 files
504 - afs (audio file sender) is now called vss (virtual streaming
505 system). Consequently, the permission flags specified in
506 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
507 become VSS_READ and VSS_WRITE respectively.
508 - para_audiod/para_filter: Fix a bug that caused the last chunk
509 of audio data not being written under certain circumstances
510 - audiod: compute the difference of server time and local time
511 correctly
512 - para_server/para_audiod: Fix some memory leaks
513 - documentation improvements
514 - configure.ac: fix checks for para_krell
515 - new man pages
516
517 -------------------------------------------
518 0.2.14 (2006-10-15) "transient singularity"
519 -------------------------------------------
520
521 The only major enhancement of this version is the osx writer which completes
522 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
523 until now for reasons beyond the scope of this changelog entry.
524
525 - new output plugin for Mac Os: the osx writer
526 - rename configure command line options from --enable-xxx-headers to
527 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
528 - configure: new command line options: --with-mad-headers,
529 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
530 - some robustness fixes
531 - dymamic audio format recognition for audiod
532 - para_server: new command line option: --autoplay_delay
533 - para_audiod: new command line option: --clock_diff_count
534
535 ---------------------------------------
536 0.2.13 (2006-07-14) "sonic convolution"
537 ---------------------------------------
538
539 A bunch of new features and core changes.
540
541 - the new paraslash scheduler, short and sweet.
542 - Support for m4a/mp4 files via the new aac audio format
543 handler/filter (requires libfaad).
544 - each writer has its own command line parser, just like
545 para_recv and para_filter.
546 - para_client and para_audioc use the error subsystem
547 - writers are integrated in para_audiod (currently linux-only)
548 - para_client is integrated in para_audiod
549 - random/playlist selector: improved info strings
550 - new audiod commands: tasks, kill
551 - update to libortp-0.10.1
552 - para_fade: wake time defaults to 8 hours from now
553 - update to autoconf-2.60
554
555 ------------------------------------------
556 0.2.12 (2006-05-12) "oriented abstraction"
557 ------------------------------------------
558
559 Many user-visible changes in this release and lots of new
560 features:
561
562
563 - the new optional dccp sender/receiver. It uses the datagram
564 congestion control protocol. You'll need a fairly new kernel
565 for this.
566 - paraslash works on Mac OS X (thanks to Gerd Becker)
567 - para_play renamed to para_write
568 - modular output plugin design (writers) for para_write
569 - new file_writer output plugin for para_write
570 - compress filter speed improvements
571 - update to libortp-0.9.1
572 - update to gengetopt-2.17rc
573 - para_client no longer depends on libreadline (as the
574 code for the interactive mode was removed).
575 - gcc-2-95 is no longer a supported compiler. It may still
576 work, but it gets no more testing.
577 - the tarball no longer contains the screenshot images which
578 reduces its size quite a bit.
579 - configure: new command line options: --enable-mysql-headers
580 and --enable-mysql-libs
581
582 ------------------------------------
583 0.2.11 (2006-03-11) "atomic duality"
584 ------------------------------------
585
586 Here it is, the first paraslash release developed with git. There
587 are fairly many user-visible changes in this release. As two out of
588 the three "database tools" of paraslash don't use a database at all,
589 they are now called "audio file selectors" instead.
590
591
592 - the cdt command (change database tool) becomes chs (change
593 selector)
594 - no more colon separators: The syntax of some options of
595 para_audiod and para_filter have changed. Use --help for
596 more info (and some examples).
597 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
598 - switch from cvs to git (should've done that earlier)
599 - the new ipc subsystem
600 - new audio file selector: playlist
601 - para_server: the dopey selector is now called "random",
602 and is the default selector. Use the --selector option to
603 choose another selector at startup, or the chs command to
604 change the selector at runtime.
605 - X86_64 fixes (thanks to Steffen Klassert)
606 - para_play fixes
607
608 --------------------------------------
609 0.2.10 (2006-02-17) "cyclic attractor"
610 --------------------------------------
611
612 Huge documentation update, a scrollable window for para_gui, ortp
613 improvements, and of course many small fixes not mentioned here.
614 The diffstat below is rather misleading as most insertions are due
615 to the new source documentation.
616
617 - autoconf cleanup
618 - para_server also uses the new error subsystem
619 - lots of new documentation (UTSL)
620 - gui improvements:
621 - keysyms for cursor keys and for next/previous page keys
622 - scrollable output window
623 - new internal commands: scroll up/down, page up/down
624 - fix color of command output.
625 - ortp: the --chunk_time and --header flags are no longer needed
626 for para_recv/para_audiod as this information is now encoded in
627 each rtp packet sent by para_server.
628
629 -------------------------------------------
630 0.2.9 (2006-01-24) "progressive turbulence"
631 -------------------------------------------
632
633 Internal audiod receivers/filters, the new error subsystem and
634 a lot of small improvements.
635
636 - para_recv and para_filter are integrated into the para_audiod
637 binary, i.e. audiod no longer spawns a new process for
638 each receiver/filter. As para_recv and para_filter might be
639 useful as standalone programs, they still get built (linked
640 against the same object files that are also used for audiod).
641 - further ortp timing improvements should reduce the CPU usage
642 of the ortp receiver.
643 - improved audio grabbing. The 'grab' command of para_audiod
644 has its own set of command line options. Read the output of
645 "para_audioc help grab" for more info.
646 - oggdec: configurable input prebuffer size.
647 - the new error subsystem gives better error diagnostics
648 and reduces code size.
649
650 -----------------------------------------
651 0.2.8 (2006-01-02) "dynamic accumulation"
652 -----------------------------------------
653
654 The new modular filter design and the para_play-hangs bugfix.
655
656 - new executable: para_filter. It combines para_mp3dec,
657 para_oggdec and para_compress. It also adds a further filter
658 type, wav, that just inserts a wave header at the desired point
659 of the filter chain. All 'piping' is done in-memory (i.e. no
660 read/write operations are used).
661 - para_play: fix a stupid bug that caused it to hang under
662 certain circumstances.
663
664 -------------------------------------------
665 0.2.7 (2006-12-27) "transparent invariance"
666 -------------------------------------------
667
668 Not many user-visible changes but a fair amount of internal improvements.
669
670
671 - The http sender buffers data if it can not be sent
672 out immediately (because the socket is not writable). This
673 should prevent para_server from shutting down the connection
674 too early on a loaded network.
675 - para_play also prebuffers data if it is told to start at a
676 future time by the --start_time option.
677 - The return of para_recv: It combines para_ortp_recv and
678 para_http_recv. Use the --receiver option to switch between
679 the two. para_recv builds without libortp, but contains
680 only the http receiver in this case.
681 - update to ortp 0.8.1. As this ortp release contains incompatible
682 changes, para_recv-0.2.7 won't link against older ortp libs.
683 - improved ortp timings.
684 - use of gcc-extensions that #define away for non-gcc and
685 gcc < 3.0.
686
687 -------------------------------------------
688 0.2.6 (2005-10-29) "recursive compensation"
689 -------------------------------------------
690
691 Transparent session encryption (uses openssl's Alleged RC4 cipher),
692 the internal find command and several other improvements and cleanups.
693
694 - Encrypt the session if encryption is requested by the client
695 (default for para_client 0.2.6). This is backwards
696 compatible, so older clients can still connect to para_server
697 0.2.6. Use the new client option --plain to request an
698 uncrypted session (off by default, must be set to on in
699 order to connect to para_server 0.2.x with 0 <= x <= 5).
700 - para_server uses an internal function to locate audio files
701 rather than calling find(1). The server option
702 --mysql_audio_file_dir replaces --mysql_find_cmd.
703 - documentation update
704 - man pages
705 - header file cleanup
706 - para_client code cleanup
707 - para_gui: faster display of output of display commands
708
709 ------------------------------------------
710 0.2.5 (2005-10-13) "aggressive resolution"
711 ------------------------------------------
712
713 This release adds internal senders, i.e. no more external programs are
714 spawned for sending out the audio data. There are two different senders
715 available: The http sender and the ortp sender (former para_send which
716 is no longer needed).
717
718 The new sender code has a plugin-like design so it can be easily
719 extended should there be be any future need for supporting another
720 network streaming protocol. All senders are completely independent of
721 each other. In particular, the http and the ortp sender can operate
722 in parallel.
723
724 - new server command: sender to control senders at runtime.
725 Read the output of "para_server -h" and "para_client help
726 sender" for more information.
727 - para_recv renamed to para_ortp_recv
728 - new executable: para_http_recv, a simple command line
729 http receiver.
730 - major afs/mp3/ogg code simplifications due to internal
731 senders.
732 - ogg timing improvements
733 - fix several minor memory leaks (found by valgrind)
734 - empty stream definitions work again
735 - com_ne(): ignore errors on remove
736 - audiod: fix segfault on server restart
737
738 ---------------------------------------
739 0.2.4 (2005-09-21) "toxic anticipation"
740 ---------------------------------------
741
742 Several small improvements, fixes and the new grab command.
743
744 - audiod:
745 - new command: "grab" to grab the output of the stream reader
746 or any filters. Read the output of "para_audioc help grab"
747 for more information.
748 - fix memory leak
749 - code cleanup
750 - audioc: new command line option: --bufsize to specify a
751 buffer size different from the default size 8192.
752 - improved error diagnostics for para_play.
753 - new configure option: --enable-ssldir so search for openssl in
754 non-standard places
755 - sdl_gui: Make it look nice again for 1024x768
756 - server: report total size of memory allocated with sbrk by malloc,
757 new command line option: --announce_time
758
759 -----------------------------------------
760 0.2.3 (2005-09-01) "hydrophilic movement"
761 -----------------------------------------
762
763 Two new executables and major feature enhancements.
764
765 - audiod filters: It is now possible to specify arbitrary many
766 (including none) filters for each supported audio
767 format. This can be used e.g. for normalizing volume,
768 transforming or grabbing the audio stream, or for using
769 visualizers. Read the output of "para_audiod -h" for the
770 syntax of the new --filter_cmd option.
771 - new executable: para_play, a tiny alsa player. It
772 can play wave files or raw pcm (16 bit little endian)
773 from stdin.
774 - new executable: para_compress, a dynamic range compressor
775 intended to keep audio output at a consistent volume. Derived
776 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
777 - audiod: New option: --stream_delay. This can be used in
778 a local network to syncronize the audio output of all
779 clients that play the same stream.
780
781 ------------------------------------------
782 0.2.2 (2005-08-19) "tangential excitation"
783 ------------------------------------------
784
785 Mostly internal changes in this release, but also some new commands
786 for the mysql database tool.
787
788 - cleanup exec.c, fix para_exec bug
789 - compile time loglevel (log messages below the given level
790 won't be compiled in, which reduces the size of the
791 resulting binaries)
792 - new log macros that shorten the size of the source code.
793 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
794 to send only zeros. With this workaround, para_audioc works
795 again.
796 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
797 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
798 (new entry), "snp" (set numplayed). Read the manual for more
799 information.
800
801 ---------------------------------------
802 0.2.1 (2005-08-15) "surreal experience"
803 ---------------------------------------
804
805 Here comes paraslash-0.2.1. It contains a couple of new features and,
806 surprise, only minor bug fixes.
807
808 - kill noisy mp3 debug message
809 - cleanup of the build system
810 - para_server and para_client directly use the crypto routines
811 of the openssl library rather than invoking the openssl command
812 line utitlity
813 - server/audiod: new option --user to switch to the given user
814 when invoked as root. Read the output of "para_server -h" for
815 more information.
816 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
817 status. Default: "para_audioc stat"
818 - sdl_gui: new option --pic_cmd to be used to download the picture.
819 Default: "para_client pic"
820 - audiod: 5 slots ought to be enough for everybody
821 - audiod: new status item: Uptime, kill hup command
822
823 ------------------------------------------
824 0.2.0 (2005-08-06) "distributed diffusion"
825 ------------------------------------------
826
827 After several month of increased development activity, paraslash-0.2.0
828 has arrived. It contains many new features and is much more
829 self-contained than the old 0.1.x series. Enjoy!
830
831
832 - para_server: fix hang on song change and crash on sighup.
833 Speed up mysql queries. The DIR_LIKE macro is gone.
834 - new executables: para_audiod, the local audio daemon that
835 starts playback (uses SCM_CREDENTIALS socket magic) and
836 para_audioc, the corresponding client.
837 - new executables: para_mp3dec/para_oggdec, two really teensy
838 decoders. para_mp3dec is based on libmad, para_oggdec requires
839 libvorbisfile.
840 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
841 they are now called para_send and para_recv respectively.
842 - documentation updates
843 - para_gui is themable. For now there is the default theme that
844 looks as before and the simple theme: blue and easy.
845 - gui: audio streaming is now handled by audiod. Time display shows
846 playback time rather than streaming time
847 - slider: update to libzmw-0.2.0
848 - para_krell: fix crash on server shutdown
849 - switch from gzip to bzip2
850
851 ----------------------------------------
852 0.1.7 (2005-04-18) "melting penetration"
853 ----------------------------------------
854
855 The main change in this release is clearly the oggvorbis rewrite,
856 but there are also lots of smaller changes. If you intend to use both
857 the mp3 and the ogg plugin, it is recommended to use software mixing,
858 e.g. the dmix plugin which is provided by ALSA.
859
860 - new executables: para_ovsend and para_ovrecv for sending/receiving
861 oggvorbis files via rtp. Requires the open rtp library. Get it at
862 http://www.linphone.org/ortp/
863 - rewrite of the ogg_vorbis core code
864 - configure detects libzmw and, if detected, includes
865 para_slider to the list of binaries to be built by make
866 - server stream writers read from their associated fifo rather
867 than from stdin
868 - slider: two new sliders, lastplayed and numplayed
869 - fix nasty double free bug which caused random segfaults in case of
870 mp3 files with invalid header information
871 - gui: new command line option: --stream_timeout=seconds to
872 deactivate a slot if it is idle for that many seconds (default=`5')
873 - diffstats
874
875 ---------------------------------------
876 0.1.6 (2005-03-05) "asymptotic balance"
877 ---------------------------------------
878
879 Only little user-visible changes in this release. Mainly bugfixes and
880 core code cleanup. This is probably the most stable version ever if you
881 stick to mp3...
882
883 - fix several memory leaks
884 - rename default name of mysql database from "music" to "paraslash".
885 Use para_server's --mysql_database option if you do not want to
886 rename your old database.
887 - rework ogg vorbis code
888 - make update command work on mysql servers with LOCAL_INFILE
889 disabled
890 - gui: improved stream I/O (slots)
891 - simplified audio format API
892 - para_pob_ogg is gone
893
894 ------------------------------------
895 0.1.5 (2004-12-31) "opaque eternity"
896 ------------------------------------
897
898 Let's slide gently into the new year.
899
900 - new: para_slider (not built automatically, type "make
901 para_slider" to build). A toy for those who always felt that
902 creating stream definitions is difficult. See screenshots,
903 README and FEATURES for more info.
904 - improved signal handling. Fixes server segfault on SIGHUP
905 for linux kernels newer than Aug 24 2004 and makes para_gui
906 race-free.
907 - reload database tool on SIGHUP
908 - improved help message for sl
909 - do not log "broken pipe" messages as errors. They are
910 perfectly ok.
911 - fix wrong error message on permission errors
912
913 -----------------------------------------
914 0.1.4 (2004-12-19) "tunneling transition"
915 -----------------------------------------
916
917 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
918 Hopefully, most of them got wiped out with this release. Some
919 enhancements went also in.
920
921 - improved error diagnostics for all commands
922 - stradd/picadd: overwrite previous contents if entry already
923 exists, rather than returning errors
924 - stradd: use current stream if invoked without args
925 - faster (and hopefully more stable) ogg-vorbis handling
926 - para_krell: reap children to avoid zombie-flooding in case
927 no server is running
928 - si: report also server pid
929 - server: don't busy-loop if dbtool reports only invalid files.
930 - gui: CTRL+C works again, fix stream_read command line option
931 - fix pic_add, hist
932 - fix mysql dbtool startup in case no database exists
933 - many small fixes and cleanups
934
935 ---------------------------------------
936 0.1.3: (2004-12-10) "vanishing inertia"
937 ---------------------------------------
938
939 Starting from this release, the database tools are integrated in the
940 server binary. This decreases server startup time, reduces code size
941 and speeds up database commands. However, the layout of the underlying
942 mysql database changed only slightly and 0.1.3 should be backwards
943 compatible in that respect.
944
945 Visible changes:
946
947 - If mysql is not detected at compile time, or fails to init
948 at runtime, fall back to the dopey database tool which should
949 always work.
950 - para_dbtool and dbtool.conf are gone. All mysql specific
951 options are read from server.conf and are prefixed by 'mysql_'.
952 - new command: cdt (change database tool)
953 - new command line option: dbtool (choose startup database tool)
954 - The name of current stream is now stored in the database,
955 so paraslash remembers its current stream when restarted.
956 - new command: csp (change stream and play)
957 - para_gui also reports current database tool and server uptime
958
959 -------------------------------------------
960 0.1.2: (2004-11-28) "spherical fluctuation"
961 -------------------------------------------
962
963 Point release before the big dbtool changes go in.
964
965 - dbtool: rename ca to cam (copy all meta data). It now also
966 copies numplayed and lastplayed time as well as the picture
967 id.
968 - fix endless-loop-bug caused by mp3 files with invalid header
969
970 -----------------------------------------
971 0.1.1: (2004-11-05) "floating atmosphere"
972 -----------------------------------------
973
974 - gkrellm plugin
975 - new dbtool command: mbox. Browse your sound-file collection
976 with your favorite mail reader.
977 - several small fixes
978
979 -------------------------------------
980 0.1.0: (2004-10-22) "rotating cortex"
981 -------------------------------------
982
983 - fix logging bug for loglevel > VERBOSE
984 - fix skip command
985 - correct timings for vbr mp3s
986 - modular audio format support
987 - ogg-vorbis support (experimental)
988 - new server option: autoplay
989
990 -----------------------------------------
991 0.0.99: (2004-07-25) "harmonic deviation"
992 -----------------------------------------
993
994 - rename projectname from icc to paraslash (play, archive, rate
995 and stream large audio sets happily)
996 - paraslash is no longer restricted to one particular audio
997 streaming software
998 - new dbtool commands (stradd, strq, strdel) for easy stream
999 managment w/o configuration file. That obsoletes stream_defs
1000 file/config option for dbtool.
1001 - picadd accepts jpeg data from stdin
1002 - new server commands: ps (select previous stream), sc (song change)
1003 - new default pictures for sdl_gui
1004 - gui: new key_map option for binding commands and internal
1005 functions to arbitrary keys, nice help screen, rip out
1006 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1007 show silly logo on startup
1008 - new executables: para_fade for fading volume, para_dbadm for
1009 manipulating attributes
1010 - cdb adds _all_ tables to mysql database
1011 - revised and beautified documentation
1012 - sample dbtool rewritten in C
1013 - autoconf
1014
1015 ---------------------------------------------
1016 0.0.98: (2003-12-26) "incremental smoothness"
1017 ---------------------------------------------
1018
1019 - kick icecast in favour of poc. That removes some races and reduces
1020 core code considerably.
1021 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1022 give finer info on afs' status.
1023 - gui can start decoder (see config options). Further new gui
1024 commands: refresh (^L), jmp (F1-F10)
1025 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1026 terminates gui. This fixes dead instances consuming memory
1027 continuously.
1028 - new dbtool command: verb for sending verbatim sql queries.
1029 - fix pid_list races (by removing pid_list)
1030 - codename funnies
1031
1032 --------------------
1033 0.0.97: (2003-10-26)
1034 --------------------
1035
1036 - installation prefix now defaults to /usr/local
1037 - new commands for gui: snozze, sleep and reread config
1038 - config file for gui and sdl_gui
1039 - fix problems with filenames containing funny characters
1040 (reported by Thomas Forell)
1041 - improved signal handling for gui, now it rereads conf on SIGHUP
1042 - new dbtool command: cdb (create database)
1043 - switch from argtable to gengetopt
1044 - major code cleanup and speed improvements
1045 - fix several potential buffer overflows
1046 - many small fixes and cleanups
1047
1048 -------------------
1049 0.0.96 (2003-08-30)
1050 -------------------
1051
1052 - easy stream_defs syntax
1053 - sdl_gui can display images associated to the file being played
1054 - Major feature enhancements for icc_gui including dynamic text
1055 placement and the top/bottom window design
1056 - vrfy/clean now also checks for NULL values in attributes as
1057 well as for invalid picture pointers
1058 - fix long outstanding case sensitivity bug
1059 - many small fixes and cleanups
1060
1061 -------------------
1062 0.0.95 (2003-06-29)
1063 -------------------
1064
1065 - sdl gui runs much faster
1066 - new dbtool command: ca (copy attributes)
1067 - count and display number of times the song has been played
1068 - new feature: scoring
1069 - command line options for sdl_gui
1070 - simpler syntax of streams file
1071 - decrease network traffic of stat
1072 - fix zombie bug
1073 - many small fixes and cleanups
1074
1075 -------------------
1076 0.0.94 (2003-05-04)
1077 -------------------
1078
1079 - new server command: ns (next stream)
1080 - new icc_gui command: c (change stream)
1081 - internal mp3info
1082 - stat shows also id3 tag info
1083 - new sdl based gui
1084 - log flodding bug fixed
1085 - many small fixes and cleanups
1086
1087 -------------------
1088 0.0.93 (2003-03-28)
1089 -------------------
1090
1091 - colors for icc_gui
1092 - icc_gui sets volume directly (linux only)
1093 - proper locking that fixes some races
1094 - fix security bug that caused commands to be executed even
1095 with unsufficient permissions
1096 - new command: hup to make all servers reread their configuration file
1097 - icecast meta data streaming
1098 - many small fixes and cleanups